Quasi-Newton Updating for Large-Scale Distributed Learning
This addresses the need for efficient distributed computing in statistical analysis, offering improvements over existing methods that require more iterations.
The paper tackles the problem of distributed learning by developing a distributed quasi-Newton (DQN) framework that reduces computation and communication complexity, and it theoretically demonstrates statistical efficiency over a small number of iterations under mild conditions.
Distributed computing is critically important for modern statistical analysis. Herein, we develop a distributed quasi-Newton (DQN) framework with excellent statistical, computation, and communication efficiency. In the DQN method, no Hessian matrix inversion or communication is needed. This considerably reduces the computation and communication complexity of the proposed method. Notably, related existing methods only analyze numerical convergence and require a diverging number of iterations to converge. However, we investigate the statistical properties of the DQN method and theoretically demonstrate that the resulting estimator is statistically efficient over a small number of iterations under mild conditions. Extensive numerical analyses demonstrate the finite sample performance.